GITHUB AGENT SKILLS / 技能收录
给你的 AI 助手
装上专业技能
我们从 GitHub 持续收录、审核并分类优质的 Agent Skills，按职业与使用场景整理，复制一行命令即可安装到 Claude Code、Cursor 等智能体。
按职业浏览 Agent Skills
为不同岗位挑选的高频技能
按使用场景浏览 Agent Skills
从你要完成的任务出发
按创作者浏览 Agent Skills
按 GitHub Stars 排序的优秀创作者
Microsoft
@microsoft
openclaw
@openclaw
OpenAI
@openai
The Apache Software Foundation
@apache
GitHub
@github
Anthropic
@anthropics
AutoGPT
@Significant-Gravitas
Garry Tan
@garrytan全部 Skills 196921
hexstrike-ai-authorized-engagement
Use when routing an authorized security assessment, bug bounty, red-team exercise, or CTF task through the local HexStrike AI runtime ; do NOT use for installation or MCP wiring (use hexstrike-ai-setup or hexstrike-ai-mcp) ; returns HexStrike AI engagement results for authorized targets only.
harness-module-authoring
Author, register, test, or review lifecycle harness modules and hook checks in agent-skills-toolchain. Use when changing ci/harness HookEvent or HookResult modules, hooks/_dispatch/manifest.json checks, fail-open behavior, redaction, persistent side effects, deduplication, calibrated-silence confidence gates, contraindication registry rows, or block-versus-inject policy.
handoff
Compact the current conversation into a handoff document for another agent to pick up.
hacktricks
Retrieve and synthesize HackTricks and HackTricks Cloud guidance for authorized security research, pentests, CTFs, hardening, incident response, reverse engineering, and web, network, Active Directory, mobile, cloud, OS, and protocol techniques. Use when the user names HackTricks or needs a concrete security technique, checklist, command, payload, prerequisite, or defensive reference. Route between the connected hacktricks MCP, the offline corpus, and the quota-limited official HackTricks AI API. Do not use it as current CVE or vendor truth, as authorization to act, or as a substitute for generic library documentation.
hacktricks-api
Query the HackTricks AI for a synthesized, curated answer when the local hacktricks corpus search is not enough. Cached (repeat queries cost nothing) and rate-limit-aware (weekly RAG quota). Covers the same broad domains as HackTricks — security, cloud, web, network, forensics, AI, and general dev. Prefer the local hacktricks skill first; use this to escalate.
grok-lane
Local Grok OpenAI-compatible lane (chenyme/grok2api bridge) - launch, keepalive, model catalog, client wiring and budget caps for grok-4.x models on this estate.
grill-with-docs
Use when stress-testing a plan against the workspace's domain knowledge and prior decisions - challenge terminology, sharpen fuzzy language, and write resolved decisions back. Use when the user says grill against our docs or decisions, or wants a glossary-aware design grill. NOT for a plain plan interview (use grill-me) or a program-scale scope (use grill-me-extensive).
grill-me
Interview the user about a plan or design, resolving each branch of the decision tree. Use when the user wants to stress-test a plan or says grill me. NOT for program-scale or under-specified scopes (use grill-me-extensive).
grill-me-extensive
Program-scale extensive grilling for wide or under-specified decision spaces. Scout context, fan out a brainstorm to surface directions beyond the stated ask, structure a categorized decision list, then walk one question at a time with a recommendation, capturing every decision. Modes: expansion, modification, scope-change, modernization, better-ideas, red-team, simplification, prioritization.
grant-proposal-architecture
Use when shaping a grant, fellowship, or project proposal into a coherent reviewer-facing argument with work packages, risk logic, and feasibility.
ghidra-mcp
Use when installing or configuring GhidraMCP for a local Ghidra workflow, or when bridging its MCP client into Codex, Claude Code, or Gemini CLI after the Ghidra extension is installed.
ghidra-bridge
Use when automating Ghidra from Python, driving analyses from a REPL or script, or exchanging program state with other reverse-engineering workflows. NOT for MCP wiring (use ghidra-mcp).
ghidra-automation-recipes
Use when automating Ghidra analysis with Script Manager scripts, ghidra_bridge, or repeatable Python-driven recipes, especially for batch queries, symbol harvesting, structure extraction, or cross-tool analysis handoffs.
gh-fix-ci
Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions; use `gh` to inspect checks and logs, summarize failure context, draft a fix plan, and implement only after explicit approval. Treat external providers (for example Buildkite) as out of scope and report only the details URL.
gh-address-comments
Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.
gem-hunt
Use when the user wants to DISCOVER high-value improvements to their OWN agentic workspace/toolkit - new hooks, behaviors, gems, skills, MCPs, or frontier-harness patterns - via a wide multi-lens agent fan-out, then a consolidated adopt/prune decision. The named "agentic-perfection sweep" / "extensive gem search". Triggers: "find more gems", "what are we missing vs other tools".
gcp-compute-foothold
Use when operating the bolt-finance-rides GCP compute primitive - minting cloud-platform SA tokens via VM metadata + serial-port read (IAM/iamcredentials SERVICE_DISABLED), standing up compute/GPU rigs on idle quota (A2 192 vCPU / A100-40GB / L4 x32 / T4 x16, europe-west1), running in-VPC workloads, or re-checking quota/API pre-run. Triggers - gcp-compute-foothold, bolt-finance-rides, SA token mint, metadata token, serial-port token, a2-highgpu, GPU rig on bolt quota, fenti OAuth redundancy.
gateguard
Fact-forcing gate that blocks Edit/Write/Bash (including MultiEdit) and demands concrete investigation (importers, data schemas, user instruction) before allowing the action. Measurably improves output quality by +2.25 points vs ungated agents.